Regarding -L, consider a line ending in unescaped white space and the next

non-empty line to form one single line, as required by XPG 4.
While here, use isblank(3) to identify white space instead of assuming
that only blank and tab are white space characters.

Geoff Clare <gwc at opengroup dot org> agrees that the standard for -L
should say "A line is considered to end with the first <newline>
unless the last character of the line is an *unescaped* <blank>;
an *unescaped* trailing <blank> signals continuation to the next
non-empty line, inclusive."  The word *unescaped* is missing from
the standard, but we regard that as a goof in the standard.

Bug noticed by, much feedback from, and "should go in now"  guenther@

2 files changed, 25 insertions, 12 deletions

diff --git a/usr.bin/xargs/xargs.1 b/usr.bin/xargs/xargs.1
index 00a76bec8e8..f2725b675dd 100644
--- a/usr.bin/xargs/xargs.1
+++ b/usr.bin/xargs/xargs.1
@@ -1,4 +1,4 @@
-.\"	$OpenBSD: xargs.1,v 1.24 2010/02/04 23:12:48 schwarze Exp $
+.\"	$OpenBSD: xargs.1,v 1.25 2010/03/25 01:03:57 schwarze Exp $
 .\"	$FreeBSD: xargs.1,v 1.30 2003/05/21 21:07:28 ru Exp $$
 .\"
 .\" Copyright (c) 1990, 1991, 1993
@@ -34,7 +34,7 @@
 .\"
 .\"	@(#)xargs.1	8.1 (Berkeley) 6/6/93
 .\"
-.Dd $Mdocdate: February 4 2010 $
+.Dd $Mdocdate: March 25 2010 $
 .Dt XARGS 1
 .Os
 .Sh NAME
@@ -170,7 +170,9 @@ Call
 .Ar utility
 for every
 .Ar number
-of lines read.
+of non-empty lines read.
+A line ending in unescaped white space and the next non-empty line
+are considered to form one single line.
 If EOF is reached and fewer than
 .Ar number
 lines have been read then
diff --git a/usr.bin/xargs/xargs.c b/usr.bin/xargs/xargs.c
index bba5d2589c5..3684c43a6b3 100644
--- a/usr.bin/xargs/xargs.c
+++ b/usr.bin/xargs/xargs.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: xargs.c,v 1.26 2009/10/27 23:59:50 deraadt Exp $	*/
+/*	$OpenBSD: xargs.c,v 1.27 2010/03/25 01:03:57 schwarze Exp $	*/
 /*	$FreeBSD: xargs.c,v 1.51 2003/05/03 19:09:11 obrien Exp $	*/
 
 /*-
@@ -38,6 +38,7 @@
 #include <sys/param.h>
 #include <sys/wait.h>
 
+#include <ctype.h>
 #include <err.h>
 #include <errno.h>
 #include <fcntl.h>
@@ -239,10 +240,23 @@ main(int argc, char *argv[])
 static void
 parse_input(int argc, char *argv[])
 {
+	int hasblank = 0;
+	static int hadblank = 0;
 	int ch, foundeof = 0;
 	char **avj;
 
-	switch (ch = getchar()) {
+	ch = getchar();
+	if (isblank(ch)) {
+		/* Quotes escape tabs and spaces. */
+		if (insingle || indouble)
+			goto addch;
+		hasblank = 1;
+		if (zflag)
+			goto addch;
+		goto arg2;
+	}
+
+	switch (ch) {
 	case EOF:
 		/* No arguments since last exec. */
 		if (p == bbp) {
@@ -252,18 +266,14 @@ parse_input(int argc, char *argv[])
 			exit(rval);
 		}
 		goto arg1;
-	case ' ':
-	case '\t':
-		/* Quotes escape tabs and spaces. */
-		if (insingle || indouble || zflag)
-			goto addch;
-		goto arg2;
 	case '\0':
 		if (zflag)
 			goto arg2;
 		goto addch;
 	case '\n':
-		count++;
+		hasblank = 1;
+		if (hadblank == 0)
+			count++;
 		if (zflag)
 			goto addch;
 
@@ -382,6 +392,7 @@ addch:		if (p < ebp) {
 		*p++ = ch;
 		break;
 	}
+	hadblank = hasblank;
 }
 
 /*